> ## Documentation Index
> Fetch the complete documentation index at: https://docs.synapsai.cloud/llms.txt
> Use this file to discover all available pages before exploring further.

# Optimizing costs

> Reduce spend with readiness levels, quantization, context tuning, and autoscaling

SynapsAI Cloud bills for active GPU memory, optional Super fast storage, and inference usage. These strategies help you balance latency and cost.

## Choose the right readiness level

| Level            | When to use                         | Cost profile                                                   |
| ---------------- | ----------------------------------- | -------------------------------------------------------------- |
| **Cold start**   | Dev, testing, rare batch jobs       | Lowest idle cost; highest first-request latency                |
| **Super fast**   | Most production workloads           | Pay only while instances run; fast load from prepped artifacts |
| **Always ready** | Lowest-latency production workloads | Highest cost — instances run 24/7                              |

See [Core concepts](/core-concepts#readiness-levels) for details. Downgrade readiness in non-production environments and use [worker timeout](/manage/autoscaling#worker-timeout) to shut down idle instances.

## Use quantization or lower precision

Quantization and lower-precision formats reduce memory usage by representing weights and activations with fewer bits. This lowers infrastructure cost and often improves throughput.

Supported methods include INT4 and INT8 quantization, as well as FP8, BF16, and EETQ where available.

### Memory and cost benefits

* **4-bit quantization (INT4):** \~75% less memory than FP16. A 33B model can run in \~24 GB VRAM; a 65B model in \~46 GB.
* **8-bit quantization or FP8 (INT8/FP8):** \~50% memory reduction with minimal quality impact; often 1.5×–2× faster inference.

Quantization also reduces storage size and cold-start load time for smaller weight files.

Configure precision and quantization during [deployment](/guides/deploy-model#quantization).

## Shorten the context window for LLMs

A shorter context window reduces VRAM needed to store activations. For example, reducing a 32k context to 16k typically cuts context memory roughly in half.

Set the context length in your model's deployment configuration to match your application's actual needs.

## Tune autoscaling and worker timeout

* Raise **scale up threshold** if you prefer fewer, busier instances.
* Lower **worker timeout** to shut down idle instances faster on Super fast and Cold start deployments.
* Use **minimum instances = 0** (where supported) for workloads that tolerate cold starts.

See [Autoscaling](/manage/autoscaling) for parameter descriptions.

## Monitor usage

Track per-model cost, token usage, and instance count on the **Analytics** tab. See [Manage models](/manage/model#analytics) and [Billing](/manage/billing).

## Billing safeguards

* Enable **auto-pay** so deployments are not interrupted when credits run low.
* Review estimated costs in the deployment UI before confirming.
* Scope [API keys](/manage/api-keys) to specific models to limit blast radius.